#46 Qwen3.5-122B-A10B
medium- Cost
- $0.019
- Time
- 48.7s
- Tokens
- 6,034 tok
AI BENCHY Compare
Qwen: Qwen3.5-122B-A10B vs StepFun: Step 3.7 Flash
Summary
The average score is effectively tied at 7.7 vs 7.7. Step 3.7 Flash (low) has the lower benchmark cost at $0.341 vs $0.588. Step 3.7 Flash (low) is faster at 15.74s vs 42.49s, with pass rates of 73.0% vs 68.3%.
Recommended modelStep 3.7 Flash (low)It has the best score here (7.7), while costing about 1.7x less than Qwen3.5-122B-A10B (medium).
Last updated at: 2026-07-14
| Metric | Qwen3.5-122B-A10B Qwen3.5-122B-A10B medium | Step 3.7 Flash Step 3.7 Flash low |
|---|---|---|
| Score | 7.7 | 7.7 |
| Rank | #46 | #51 |
| Reliability | 10.0 | 10.0 |
| Consistency | 8.8 | 8.4 |
| Tests Correct | ||
| Attempt pass rate | 73.0% | 68.3% |
| Flaky tests | 3 | 4 |
| Total Runs | 63 | 63 |
| Cost per result | 5.235 | 2.840 |
| Total Cost | $0.588 | $0.341 |
| Input Price | $0.260 / 1M | $0.200 / 1M |
| Output Price | $2.080 / 1M | $1.150 / 1M |
| Total Input Tokens | 41,832 | 40,101 |
| Output Tokens | 26,187 | 289,325 |
| Reasoning Tokens | 251,028 | 0 |
| Response Time (avg) | 42.49s | 15.74s |
| Response Time (max) | 168.16s | 124.75s |
| Response Time (total) | 892.30s | 330.63s |
Generation showcase
Hamster playing table tennis
Prompt: Create a detailed SVG illustration of a hamster playing table tennis.
#51 Step 3.7 Flash
low
Invalid SVG
- Cost
- $0.004
- Time
- 25.3s
- Tokens
- 3,072 tok
Top Models by Score
Score vs Total Cost
Response Time (avg)
Score vs Response Time (avg)
Total Output Tokens
Score vs Total Output Tokens
Category Breakdown
| Coding | Score | Consistency | Attempt pass rate | Flaky tests | Tests Correct | Response Time (avg) | Input Tokens | Output Tokens | Reasoning Tokens |
|---|---|---|---|---|---|---|---|---|---|
| Qwen3.5-122B-A10B | 6.0 | 7.2 | 55.6% | 1 | 114.48s | 7,630 | 8,057 | 82,578 | |
| Step 3.7 Flash | 8.2 | 7.2 | 88.9% | 1 | 9.46s | 7,437 | 18,685 | 0 |
Quick Compare
Switch Comparison Pair
GPT-5.6 LunahighvsStep 3.7 FlashlowClaude Opus 4.8lowvsQwen3.5-122B-A10BmediumClaude Opus 4.6mediumvsStep 3.7 FlashlowStep 3.7 FlashlowvsGrok 4.3mediumKAT-Coder-Pro V2.5highvsStep 3.7 FlashlowGPT-5.6 TerralowvsQwen3.5-122B-A10BmediumGPT-5.6 LunahighvsQwen3.5-122B-A10BmediumKimi K2.6mediumFree AvailablevsStep 3.7 FlashlowGemini 3.1 Flash LitemediumvsStep 3.7 FlashlowKAT-Coder-Pro V2.5highvsQwen3.5-122B-A10BmediumQwen3.5-122B-A10BmediumvsGrok 4.5lowQwen3.5 Plus 2026-04-20mediumvsStep 3.7 Flashlow